# GPT-5 Mini vs Grok 4.3

> Grok 4.3 is the stronger model overall, scoring 43.8 to 41.8 on the Noometry Index. GPT-5 Mini costs 2.3× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-mini-vs-grok-4-3
- Last updated: 2026-10-10
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. GPT-5 Mini scores higher in 4 categories and Grok 4.3 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.3 leads 35.9 to 23.9.
- The biggest single-benchmark swing is GPQA Diamond: 75% for GPT-5 Mini and 88.8% for Grok 4.3.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 400K.

## Snapshot

| | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.8 | 43.8 |
| Rank | 128 | 86 |
| Context | 400K | 1M |
| Input $/M | $0.25 | $1.25 |
| Output $/M | $2 | $2.50 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5 Mini: 40.1 (#146)
- Grok 4.3: 41.6 (#121)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| SciCode | 39.2% | 47.3% |
| WeirdML | 52.7% | 49.9% |
| LMArena Coding | 1406 | 1415 |
| ALE-Bench | 799.77 | 944.17 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| LMArena WebDev | — | 1357 |
| SWE-bench Multilingual | 39.7% | — |
| AlgoTune | 1.38 | — |

## Agentic & Tool Use

- GPT-5 Mini: 31.1 (#70)
- Grok 4.3: 27.7 (#99)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| Vending-Bench 2 | -31.18 | 35.26 |
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |

## Reasoning

- GPT-5 Mini: 23.9 (#168)
- Grok 4.3: 35.9 (#68)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| CritPt | 0% | 8% |
| Chess Puzzles | 30% | 25% |
| LMArena Hard Prompts | 1380 | 1396 |
| DTBench | 80.5% | 90.7% |
| LMCA | 34.2% | 38.3% |
| Epoch Capabilities Index | 145.52 | 149.16 |
| ForecastBench | 61 | 60.3 |
| ARC-AGI-2 | 4.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| NYT Connections (extended) | — | 55.2% |
| ARC-AGI-1 | 54.3% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |

## Math

- GPT-5 Mini: 46.7 (#69)
- Grok 4.3: 46.0 (#74)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.7% | 42.8% |
| FrontierMath Tier 4 | 12.2% | 14.6% |
| OTIS Mock AIME 2024-2025 | 86.7% | 93.3% |
| ProofBench | 9% | 11% |
| LMArena Math | 1378 | 1388 |
| Omni-MATH | 72.2% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |

## Knowledge

- GPT-5 Mini: 45.6 (#86)
- Grok 4.3: 52.5 (#62)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 75% | 88.8% |
| SimpleQA Verified | 21.6% | 33.2% |
| LMArena Expert | 1379 | 1385 |
| Humanity's Last Exam | 19.4% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |

## Multimodal

- GPT-5 Mini: 35.6 (#85)
- Grok 4.3: 31.6 (#104)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| LMArena Vision | 1202 | 1229 |
| VPCT | 40.2% | — |
| Blueprint-Bench 2 | — | 0% |

## Multilingual

- GPT-5 Mini: 48.9 (#137)
- Grok 4.3: 50.5 (#120)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1363 | 1385 |
| LMArena Chinese | 1385 | 1422 |
| LMArena French | 1386 | 1412 |
| LMArena German | 1366 | 1395 |
| LMArena Japanese | 1341 | 1379 |
| LMArena Korean | 1308 | 1356 |
| LMArena Russian | 1362 | 1399 |
| LMArena Spanish | 1355 | 1398 |

## Instruction Following

- GPT-5 Mini: 76.2 (#46)
- Grok 4.3: 72.1 (#140)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1357 | 1366 |
| IFEval | 92.7% | — |

## Long Context

- GPT-5 Mini: 41.9 (#132)
- Grok 4.3: 42.5 (#123)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1355 | 1393 |
| Fiction.LiveBench | 69.4% | — |

## Writing & Preference

- GPT-5 Mini: 55.2 (#148)
- Grok 4.3: 58.5 (#118)

| Benchmark | GPT-5 Mini | Grok 4.3 |
|---|---|---|
| LMArena Text | 1373 | 1397 |
| LMArena Creative Writing | 1325 | 1380 |
| LMArena Multi-Turn | 1363 | 1406 |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
| EQ-Bench 4 | — | 1075 |

## FAQ

### Is GPT-5 Mini better than Grok 4.3?

Grok 4.3 is the stronger model overall, scoring 43.8 to 41.8 on the Noometry Index. GPT-5 Mini costs 2.3× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.

### Which is cheaper, GPT-5 Mini or Grok 4.3?

GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.

### Is GPT-5 Mini or Grok 4.3 better for coding?

Grok 4.3 scores higher on coding benchmarks: 41.6 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.3 does, with 1M tokens against 400K.

### How many benchmarks do GPT-5 Mini and Grok 4.3 share?

34 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok 4.3 has 40.